How would the European Commission's new directive change AI-related liabilities?

Why is the AILD necessary?

Artificial intelligence has developed in a rapid pace during recent years, bringing many new solutions to both business operations and to society overall. However, the rise of AI has also brought with it concern relating to the damage caused by autonomous systems. The operation of AI systems can be almost completely autonomous, and their decision-making is often based on complex and opaque processes. Therefore, the solutions made by AI can come as a complete surprise even to their manufacturers and users. As more and more demanding tasks are assigned to AI systems, the risk of serious and unpredictable damage increases.

Current national liability rules are not entirely suited to handling liability claims for damage caused by AI-enabled products and services. This is because the special characteristics of AI technology, including complexity, autonomy, and opacity, may make it difficult for victims to prove the requirements for a successful liability claim. Currently, liability for compensation generally requires both the fault of the defendant as well a causal link between the defendant’s conduct and the end result. In the case of highly autonomous machines, where human intervention has been minimal or non-existent, it can be very difficult or prohibitively expensive for victims to prove these requirements.

AILD will ease the burden of proof in AI-related claims

The EU commission aims at tackling these challenges by presenting two new features in the AILD. Firstly, the directive includes a provision that improves the claimant’s right to access…